Aircraft Pneumatic Duct Wall severely dented/creased: methods/techniques to restore roundnesss???? 1

[WKTaylor]

All.. help...

Pneumatic Duct [300-PSIG air, 800F]: Inconel 625,'multiple s-turns' duct, 3.0-inch OD, 0.035-inch WT, dents creases due to miss-handling up-to ~0.20 deep. Some of the dents/creases have sharp features [very tight radii at dent center/crease-centerline].

Ideas for restoring tubing roundness? These ducts are $30K/each... last production run... now out-of-production.

What tooling works best and what is currently available? Vendors? Home-made?

Pull/push-thru ball mandrels?

Localized internal [hydraulic?] expansion tools?

High powered magnetic dent-remover [concave outer electromagnet, convex inner permanent/steel?

Regards, Wil Taylor

Trust Me! I'm an engineer!

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.

Unfortunately, in science what You 'believe' is irrelevant û "Orion"

 

[btrueblood]

aaaand...shortly after hitting "post" I recalled one other technique I have seen used, with mixed results.

If the material has ductility at low temperatures, you can use the expansion of freezing ice (or some other liquids, I think bismuth and other low mp alloys also expand on freezing - I know plutonium does, but that might not be practical for you) to pop small dents, even repeatedly freezing/thawing to slowly "ratchet" the dent out. In your case, you would want to rig some expansion plugs into the pipe to trap the water/liquid where you want it, then dip the thing into liquid nitrogen, or roll it into a deep freeze.

 

[berkshire]

Ok,
The unit I saw was actually a conversion that used the screw type expander and replaced it with a Greenlee type hydraulic pulling cylinder.
They say it is capable of expanding 12 gauge mild steel fence posts. This can start as the basis for a custom head to get into a tube and reform it by gently stretch forming the tube.
B.E.

You are judged not by what you know, but by what you can do.

 

[btrueblood]

The cylinder in the picture shown by Berkshire has a side port, and the cylinder itself is larger in diameter than the expander...so getting it into wk's tube might not work. But otherwise, yes, that's the right approach. Ideally, the joint between the cylinder and expander body would also have a pin joint, allowing the assembly a bit more flexibility to fit through a pipe bend.

The next problem to solve is what do you do about non-circular section ducts...

 

[WKTaylor]

btb...

The expander head(s) the local tube shop made have the ratio/appearance of itty-bitty "beer kegs", IE: noticibly convex barrel '(_)' shape, length ~= unexpanded barrel diameter. The hydraulic hose port is in the flat base of the 'keg' [axis]. The stubby convex barrel shape allows the expander head to pass thru fairly tight elbows. OH yeah: the Expander head is made of steel; the techs places a powerful permanent magnet on the OD to locate the center of the barrel in the duct for relatively precise positioning [obviously works for non-magnetic tubing alloys only]. during expansion the mech keeps his eyes/hand on the dent site to 'see/feel' how the process is going.

The various tube expanders mentioned by others, all appear to be intended for end-of-tube-expansion use only.

Regards, Wil Taylor

Trust Me! I'm an engineer!

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.

Unfortunately, in science what You 'believe' is irrelevant û "Orion"

 

[berkshire]

Trueblood,
There are other cylinders that can be used on a Portapower type hand pump, including some that have end fittings to get down inside a tube.
I have some alligator jaw type body jack fittings that could also be modified to spread a pipe by fitting curved ends, that could be used on non circular sections.
This is precisely what Wills crew is doing. It sounds like right now, that Wills crew has a handle on the problem, and that they need to fabricate more custom machines for the parts they cannot handle yet.
B.E.

You are judged not by what you know, but by what you can do.

 

[WKTaylor]

berkshire... I think You are right.

The shop techs who've been around awhile are worth their weights in gold. They can/have come up with some amazing 1-off tools for solving problems like this.

DANG. Too bad we [engineers] weren't clued in earlier to prevent scrapping some ducts that probably could have been salvaged.

Off-topic note.
During my general research on CRES and HRA duct failures/repairs I ran across a report that catagorized high temp duct primary-causes for failure. Crazy... I have seen each of these 'failure modes' in my career as a field service engineer.

1. Non-spec welds/weld-beads. IE: too big, too small, erattic width, re-entrant, undercutting, poor penetration, dirty/contaminated welds due to poor pre-cleaning practices and/or moisture, and/or inadequate inert gas coverage, etc.

2. Duct manufacturing [alignment] errors that lead to poor "natural fit". IE: built-in misalignment that causes mechanics to force [pre-stress] these sections to fit. In some cases this means over-stressing no-aligned flanges and-or severe movements of flexible joints [bellows, etc] to make it fit.

3. Inadequate flexibility and/or restraint design practices. IE: too stiff and/or too flexible for internal pressure, thermal, inertia [mass-flow] loads/strains; and/or mating structure flexure/strain.

4. Maintenance abuse. Mishandling before installation, mishandling during installation/removal; poor storage practices; inadvertant use as hand-hold ['chinning bar'], etc when installed; or impact/contact during other maintenance [dropped tools or trapped FOD, etc]; or failure to follow established tech data to the letter.

NOTE.
In one case, the tech data had very specific sequences to follow in order to to remove, then re-install or replace, certain duct sections. The mechs were struggling to replace one of the sections that was found cracked. The mechs had removed it, and were trying to install the replacement, without even taking a look at the tech data/procedures. Restarting the process 'by-the-book', with some variations due to the disassy circumstances, finally allowed the section to fit 'adequately'.

5. Unanticipated operating environment; or operating environment that becomes more severe over time; or a system failure up/down stream that leads to unanticipated consequences.

NOTE.
The acft I work on have all been re-engined. The new engines supply bleed air at roughly the same pressure/temperature/flow... but are able to supply this bleed air continuously; as opposed to discontinuously... due to much higher re-engine thrust ratings.

NOTE. In one acft I worked on, a duct system failure down stream caused a dramatic rise in temperature [350F up-to 650F] up-stream of the break, since the bleed-air was passing thru the up-stream pre-coolers way-too-fast and couldn't reduce the temperature as designed. This extremely hot air did very bad things to wire harnesses and aluminum structure.

Regards, Wil Taylor

Trust Me! I'm an engineer!

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.

Unfortunately, in science what You 'believe' is irrelevant û "Orion"

 

[Tmoose]

https://www.youtube.com/watch?v=cSHD9tI2vWM

I can attest this method works pretty slick on dented and even creased dirt bike exhaust ( < .06 inch thick mild steel rolled cones and "pipe" ).

In addition I prefer to have a method to modulate (release, mostly) the pressure as well. I think it helps me avoid bulges way better than just pulling the torch away.

I have no clue about the metallurgical effects of heating your material to the point its yield strength is lowered.
Legend holds, and my experience confirms, that a mild steel pipe repaired this way will dent somewhat easier the next time it and the rock meet.

 

[berkshire]

Wktaylor,
it occurs to me that while you are wrestling with damaged ducts, your bellows may be suffering the same fate.
Here is a company that manufactures bellows and expansion joints, that you may be able to work with. Whether or not they will do Milspec work you will have to find out.

You are judged not by what you know, but by what you can do.

 

[tbuelna]

Good point about the metal bellows berkshire. Most people think that since metal bellows are designed to flex they are immune to being over-stressed. Often metal bellows are only designed to handle stresses produced from a very limited amount of compression/tension, bending and torsion. The duct company I worked at designed and manufactured their own metal bellows. They were very highly engineered devices and were not simple to manufacture. For example, the convoluted bellows itself was constructed from multiple thin layers of metal rather than a single layer. Many aircraft bellows, like those that might be used on wktaylor's high temp/pressure bleed air ducts, might exceed stress limits if they are displaced just a modest amount beyond what they are designed for. The bellows wall must be thin enough to flex without applying too much force, but must also be thick enough to handle the duct pressure without bursting or ballooning.

I recall that the bellows were constrained in a fixture during welding that preloaded it and aligned the ends. If there is a need to cut a full section out of the duct and weld in a new piece, the people doing the repair will need to make sure the duct is fixtured accurately.

 

[rb1957]

whilst restoring the shape is a good thing, what about the strength properties ?

another day in paradise, or is paradise one day closer ?